Upregulation of miR-24 is associated with a decreased DNA damage response upon etoposide treatment in
highly differentiated CD8 + T cells sensitizing them to apoptotic cell death
Stefan Brunner,
1Dietmar Herndler-Brandstetter,
1Christoph R. Arnold,
1Gerrit Jan Wiegers,
2Andreas Villunger,
2Matthias Hackl,
3Johannes Grillari,
3Marı´a Moreno-Villanueva,
4Alexander Bu¨rkle
4and Beatrix Grubeck-Loebenstein
11Immunology Division, Institute for Biomedical Aging Research, Austrian Academy of Sciences, Innsbruck, Austria
2Division of Developmental Immunology, Biocenter, University of Innsbruck, Innsbruck, Austria
3Department of Biotechnology, Aging and Immortalization Research, University of Natural Resources and Applied Life Sciences, Vienna, Austria
4Molecular Toxicology Group, University of Konstanz, Konstanz, Germany
Summary
The life-long homeostasis of memory CD8+T cells as well as persis- tent viral infections have been shown to facilitate the accumula- tion of highly differentiated CD8+CD28)T cells, a phenomenon that has been associated with an impaired immune function in humans. However, the molecular mechanisms regulating homeo- stasis of CD8+CD28)T cells have not yet been elucidated. In this study, we demonstrate that the miR-232427 cluster is up-regu- lated during post-thymic CD8+T-cell differentiation in humans.
The increased expression of miR-24 in CD8+CD28)T cells is associ- ated with decreased expression of the histone variant H2AX, a protein that plays a key role in the DNA damage response (DDR).
Following treatment with the classic chemotherapeutic agent eto- poside, a topoisomerase II inhibitor, apoptosis was increased in CD8+CD28)when compared to CD8+CD28+T cells and correlated with an impaired DDR in this cell type. The reduced capacity of CD8+CD28)T cell to repair DNA was characterized by the auto- mated fluorimetric analysis of DNA unwinding (FADU) assay as well as by decreased phosphorylation of H2AX at Ser139, of ATM at Ser1981, and of p53 at Ser15. Interleukin (IL)-15 could prevent etoposide-mediated apoptosis of CD8+CD28)T cells, suggesting a role for IL-15 in the survival and the age-dependent accumulation of CD8+CD28)T cells in humans.
Key words: aging; DNA damage; DNA repair; senescence;
CD8+T cells; apoptosis; etoposide.
Introduction
Human aging is associated with the involution of the thymus (Steinmann et al., 1985), thereby narrowing the T cell repertoire in old age. While the number of naive T cells decreases, memory and effector T cells accumu late (Arnoldet al., 2011). These changes are particularly pronounced in the CD8+T cell pool (Sauleet al., 2006). The increase in the number of highly differentiated CD8+CD28) effector T cells has attracted more attention over the last decade. This specific cell type dominates the reper toire in elderly persons and is especially high in elderly persons who also have persistent infection with cytomegalovirus (CMV) (Brunneret al., 2011). High numbers of CD8+CD28)T cells are believed to be deleterious for elderly persons, as their accumulation has been linked to a shortened live expectancy, enhanced progression of age related diseases such as Alzheimer’s disease and atherosclerosis as well as with a low efficacy of vaccination (Goronzyet al., 2001; Saurwein Teisslet al., 2002; Wikby et al., 2005). For these reasons, therapeutic elimination of this particular cell population has been considered (Pawelecet al., 2006). However, potential treatment regimes to reach this goal depend on a clear under standing of the way how CD8+CD28)T cells are generated, survive, and are eliminatedin vivo. Presently, there is more controversy on these issues. While some claim that terminally differentiated T cells are particu larly prone to undergo apoptosis and are therefore short lived, others suggest that they are long lived, probably due to an intrinsic resistance to apoptosis inducing stimuli (Spauldinget al., 1999; Effroset al., 2005;
Wallaceet al., 2011).
Recent results from our laboratory demonstrate that the microRNA (miRNA) expression pattern differs between CD8+CD28+ and CD8+ CD28)T cells (Hacklet al., 2010). Of particular interest in this context is a decreased expression of the polycistronic miRNA cluster miR 1792 in CD8+CD28)T cells, which was also found to be down regulated in other human cell types following multiple rounds of division. Of note, this miR NA cluster is known to target the cyclin dependent kinase inhibitor p21 as well as the pro apoptotic B cell lymphoma 2 (BCL2) family protein Bim, critical for lymphocyte homeostasis (van Haaften & Agami, 2010). miR NAs might therefore be of importance for replicative exhaustion by inter fering with important mediators of the cell cycle and apoptosis (Grillari et al., 2010). In the course of our previous experiments, another miRNA cluster was noticed to be overexpressed in CD8+CD28)T cells, but not in the other cell types studied.
This miRNA cluster, miR 232427, has been shown to target mole cules important for DNA repair (Lalet al., 2009; Srivastavaet al., 2011).
Decreased DNA repair during cellular senescence as well as in parallel to T cell differentiation has previously been reported (Scarpaciet al., 2003;
Seluanovet al., 2004).
We therefore wondered whether the DNA damage response (DDR) might be impaired in CD8+CD28) T cells, leading to an increased susceptibility to apoptosis in this cell type. To investigate this possibility, we further analyzed miRNA expression profiles, DNA repair, and apopto sis in CD8+CD28+and CD8+CD28)T cells. We noted decreased expres Correspondence
Professor Beatrix Grubeck Loebenstein, Institute for Biomedical Aging Research, Austrian Academy of Sciences, Rennweg 10, 6020 Innsbruck, Austria.
Tel.: +43 512 583919 10; fax: +43 512 583919 8; e mail: beatrix.grubeck loebenstein@oeaw.ac.at
Accepted for publication2 March 2012
Re use of this article is permitted in accordance with the Terms and Condi tions set out at http://wileyonlinelibrary.com/onlineopen#OnlineOpen Terms
ª2012 The Authors
Aging Cellª2012 Blackwell Publishing Ltd/Anatomical Society of Great Britain and Ireland 579
Aging Cell(2012)11,pp579 587 Doi: 10.1111/j.1474 9726.2012.00819.x
Aging Cell
sion of the histone H2A family member X (H2AX), a validated target of miR 24 (Lalet al., 2009; Srivastavaet al., 2011), in CD8+CD28)T cells.
This impairment was associated with changes in DDR signaling events, with persisting DNA strand breaks (DSBs), as well as with an increased occurrence of apoptosis in CD8+CD28)T cells following induced DNA damage that could be prevented by interleukin (IL) 15. We therefore propose that the increased vulnerability of CD8+CD28)T cells to apop totic cell death is balanced by IL 15, supporting their survival in IL 15 rich niches such as the bone marrow (BM) (Herndler Brandstetter et al., 2011a,b).
Results
miRNA expression profiling of CD8+T-cell subsets
We first compared the global miRNA expression profiles of CD8+CD28+ and CD8+CD28)T cells. A log2corrected heatmap of all 21 regulated miRNAs is depicted in Fig. 1A. This screening approach revealed a pro nounced dysregulation of two miRNA clusters in CD8+CD28)T cells that is consistent over a 16 h period of cell culture (second column in the heat map), a control used to test the effect of cell culture conditions on miRNA expression. These miRNA clusters are miR 1792 and miR 232427, the members of which are cleaved out of a polycistronic pri miRNA vari ant. Compared to CD8+CD28+T cells, CD8+CD28)T cells express less miR 17, miR 19b, miR 20a, miR 92b, and miR 106a, all of which are members of the paralogous miRNA clusters miR 1792, miR 106a363, and miR 106b25 with extensive sequence homologies. This miRNA cluster is not only down regulated in highly differentiated CD8+CD28) compared to CD8+CD28+T cells, but also in other models of replicative or chronological aging (Hacklet al., 2010).
The second miRNA cluster, which is differentially expressed in CD8+CD28)T cells, is the miR 232427 cluster. Its members were up regulated in CD8+CD28)in comparison with CD8+CD28+T cells by fac tors between 1.5 and 2.9. The regulation of each miRNA variant in this
cluster was statistically significant (Padj.£0.001). We were particularly interested in the regulation of miR 24, as it targets the H2AX, an impor tant mediator of DSB repair (Lalet al., 2009; Srivastavaet al., 2011). We therefore validated our array data on miR 24 by quantitative RT PCR (Fig. 1B). We also analyzed H2AX protein expression in the two CD8+ T cell subsets. Densitometric evaluation of Western Blot data, as shown in Fig. 1C, demonstrates a decreased H2AX protein expression in CD8+CD28)T cells in comparison with their CD8+CD28+T cell counter parts. To confirm H2AX as a cellular target of miR 24, we overexpressed miR 24 in the lymphoblast T cell line Jurkat E6.1. Forty eight hour post transfection miR 24 levels were strongly up regulated (Fig. 2). This led to a significant down regulation of H2AX protein compared to controls 96 h post transfection (Fig. 2).
DDR signaling in CD8+T-cell subsets
In a next step, we analyzed whether the reduced expression of H2AX in CD8+CD28)T cells is associated with changes in the DDR in highly differ entiated CD8+T cells. As a model to induce DNA damage, isolated CD8+ T cell subsets were exposed to etoposide, a potent topoisomerase II inhib itor that interferes in the biphasic reaction of this ubiquitous enzyme and prohibits the re ligation of previously introduced DSBs (Baldwin & Osher off, 2005). As one of the earliest events in the DDR pathway is the phosphorylation of H2AX at Ser139 (cH2AX), we performed immunoflu orescence staining of cH2AX on etoposide treated CD8+ CD28+and CD8+CD28)T cells. Our results demonstrate intact H2AX phosphorylation andcH2AX cluster formation in nuclei of CD8+CD28+T cells, while CD8+CD28)T cells failed to mount such a response properly, both in terms of the percentage of cells withcH2AX clusters in their nuclei as well as the numbers of clusters in the nuclei of individual cells (Fig. 3). These results indicate an impaired DDR in highly differentiated CD8+T cells. We next elucidated the regulation of important signaling events in the cascade of the DDR. Therefore, phosphorylated (Ser1981) ataxia telangiectasia mutated (ATM) protein, one of the first sensors of
(A) (B) (C)
Fig. 1(A) Global miRNA expression profile of human CD8+CD28+and CD8+CD28 T cells. CD8+T cell subsets were analyzed directlyex vivoor cultured for 16 h:
consistently regulated miRNAs under each experimental condition are shown on a log2scale (yellow = up regulated, blue = down regulated in CD8+CD28 compared to CD8+CD28+T cells). The heatmap represents mean ratios of four experiments with cells from individual donors; mean log2regulation; FDR adjustedP£0.05; CD8+CD28 versus CD8+CD28+. (B) Quantification of miR 24 levels in CD8+CD28+and CD8+CD28 T cells by RT PCR. The bar graph shows the relative miR 24 expression normalized to GAPDH; mean fold ratios ± SEM;n= 5; **P£0.01; CD8+CD28 versus CD8+CD28+. (C) H2AX expression in CD8+CD28+and CD8+CD28 T cells. The histone H2A variant H2AX was assessed in freshly isolated CD8+T cell subsets by Western Blot analysis of which one representative example of five is shown. The bar graph shows the densitometrical evaluation of grayscale values normalized to GAPDH; mean fold ratios ± SEM;n= 5; **P£0.01; CD8+CD28+versus CD8+CD28 .
DNA damage, phosphorylated (Ser15 and Ser46) tumor protein p53 (p53), and meiotic recombination 11 (MRE11) protein, a member of the MRN nuclease⁄helicase complex, were analyzed before, during, and after etoposide induced DNA damage. To exclude a potential influence of the chronological age of individual donors, this experiment was performed on cells from young and old donors. Our results demonstrate a decreased production of MRE11 as well as a decreased phosphorylation of ATM at Ser1981 and of p53 at Ser15 in CD8+CD28)T cells following etoposide induced DNA damage, while total ATM and p53 concentrations did not differ between the two CD8+T cell subsets. Interestingly, p53 phosphory lation at Ser46 was not changed in CD8+CD28+ T cells and even increased in CD8+CD28)T cells after treatment. Identical results were obtained with cells from young and old donors demonstrating that the reduced DDR response is a general property of CD8+CD28) T cells (Fig. 4A; no statistically significant difference between the age groups;
P> 0.05). Data from young and elderly persons were combined for the
graph shown in Fig. 4B. A schematic representation of events involved in the assembly and spreading of the DSB repair complex is depicted in Fig. 4C (Sengupta & Harris, 2005; West & van Attikum, 2006).
DNA damage and repair in CD8+T-cell subsets
Next, DNA damage and the efficiency of DNA repair was analyzed in CD8+CD28+and CD8+CD28)T cells using an automated version of the FADU assay (Fig. 5). Isolated CD8+ T cell subsets were either left untreated (Control) or exposed to etoposide for 60 min (Etoposide) as well as given time to repair the induced DNA damage by removing eto poside for 15, 30, 45, 60, or 75 min before analysis. Even without treat ment, CD8+CD28) T cells displayed more DNA damage than CD8+CD28+controls. Exposure to etoposide potently induced DSBs in both cell types but the recovery time necessary to accomplish DNA repair was different in CD8+CD28+ and CD8+CD28) T cells. While CD8+CD28+ T cells recovered in < 30 min, CD8+CD28) T cells never reached their respective basal fluorescence level. Taken together, our results indicate that CD8+CD28)T cells have acquired more DSBsin vivo and have an impaired DNA repair capacity following etoposide treat mentin vitro.
Etoposide-induced apoptosis in CD8+T-cell subsets
As an impaired DDR in CD8+CD28)T cells may also lead to increased rates of apoptosis following etoposide induced DNA damage in this cell type, we cultured PBMCs containing CD8+CD28+and CD8+CD28)cells with or without etoposide for 48 h. Apoptosis was assessed by Annexin V⁄7 aminoactinomycin D (7 AAD) staining and FACS analysis with the respective subpopulations gated. CD8+CD28)T cells had significantly higher rates of apoptosis following exposure to etoposide than their CD8+CD28+counterparts (Fig. 6A). To exclude a potential influence of other cell types within the PBMC fraction, this experiment was, with iden tical results, also performed on isolated CD8+CD28+ and CD8+CD28) T cell subsets (data not shown).
To test whether the homeostatic cytokine IL 15 could rescue CD8+CD28)T cell subsets from programmed cell death, etoposide trig gered apoptosis was analyzed in the presence of IL 15. IL 15 decreased etoposide induced apoptosis in both CD8+CD28+ and CD8+CD28) T cells, resulting in a similar rate of apoptosis in both T cell subsets. IL 15 treatment caused an up regulation of the anti apoptotic molecule BCL2
like 1 (BCLXL), thereby preventing a cascade of signaling events that cul minates in the cleavage of poly (ADP ribose) polymerase 1 (PARP1) by active caspase 3 (Fig. 6B,C). CD8+CD28) T cells displayed the same defects upon etoposide treatment in the altered kinetics of ATM and p53 phosphorylation in the absence and presence of IL 15 (n 5;P> 0.05;
IL 15 treated vs. untreated cells; data not shown) confirming that IL 15 did not affect DDR signaling events.
Discussion
As CD8+CD28)T cells have short telomeres and fail to undergo substan tial proliferation following antigenic contact, this highly differentiated T cell subset has been attributed a state of replicative exhaustion and cel lular senescence (Effroset al., 2005). CD8+CD28)T cells fail to produce IL 2, are frequently oligoclonal, pro inflammatory, and are believed to take up more immunological space, which may limit the propagation and long term survival of other T cell specificities (Saurwein Teissl et al., 2002). Little is still known about the molecular mechanisms that regulate survival or death of this specific cell type.
(A)
(B)
Fig. 2(A) miR 24 overexpression in the human leukemic T cell lymphoblast line Jurkat E6.1. Cells were either left untreated (Control) or transfected with empty (Mock), nonsense miRNA (Scrambled), or miR 24 Mimic filled transfection complexes. The bar graph shows the relative miR 24 regulation normalized to GAPDH 48 h post transfection; mean log2ratios ± SEM;n= 4; ***P£0.001;
miR 24 Mimic versus controls. (B) H2AX protein expression in miR 24 transfected Jurkat cells 96 h post transfection. The H2AX content was assessed by Western Blot analysis of which one of four representative images is shown. The graph shows the densitometrical evaluation of grayscale values. Data represent fluorescence in individual lanes as percentages of total fluorescence in the whole blot;
means ± SEM;n= 4; *P£0.05; miR 24 Mimic versus controls.
miRNA have recently been recognized to play an important role in the fine tuning of a great number of immune regulatory responses (Baltimore et al., 2008) and have also been shown to be altered in several types of
‘senescent’ cells, cells which have undergone many rounds of division, among others CD8+CD28) T cells (Grillari & Grillari Voglauer, 2010;
Hacklet al., 2010). Thus, the miRNA cluster miR 1792, which targets important cell cycle regulators such as p21, has been shown to be down regulated in these cell types (Ivanovska et al., 2008). We now demonstrate that the miR 232427 cluster is up regulated in CD8+CD28)T cells. Recent miRNA studies on human leukemia cell lines, namely K562 cells, differentiated to megakaryocytes or erythrocytes and HL60 cells differentiated to macrophages or monocytes provided compel ling evidence for an up regulation of miR 24 that leads to a down regula tion of its target H2AX, consequently suppressing efficient DNA repair (Lalet al., 2009). In view of these data on other types of highly differenti ated cells, we hypothesized that the DDR of highly differentiated T cells might also be impaired.
We first analyzed whether the expression of total H2AX was altered, and did indeed find that this molecule was reduced in CD8+CD28)T cells when compared to CD8+CD28+T cells. To confirm that miR 24 can effec tively target and suppress H2AX formation, we additionally overexpressed miR 24 in the human T cell line Jurkat E6.1 and monitored the consecu tive down regulation of H2AX. To be able to study whether changes in miR 24 and H2AX expression were accompanied by alterations of DDR signaling events in CD8+CD28)T cells, DNA damage was induced in iso lated CD8+T cell subsets by etoposide, a classic chemotherapeutic agent that targets topoisomerase II. Indeed, CD8+CD28)T cells failed to phos phorylate H2AX effectively and to subsequently generate nuclearcH2AX clusters. Impaired DDR signaling in CD8+CD28)T cells was further con firmed by decreased phosphorylation of ATM (Ser1981) and of p53 (Ser15) as well as by decreased expression of MRE11 following etoposide treatment, a phenomenon that was independent of the chronological age of the donor and thus a manifestation of the intrinsic property of the respective T cell population. The fact that p53 phosphorylation at Ser46 was not changed in CD8+CD28+ T cells and even increased in CD8+CD28)T cells after treatment demonstrated that DDR signaling was really reduced in comparison with CD8+CD28+T cells.
To provide definite evidence that there is a connection between miR 24 and the reduction of DDR signaling in CD8+CD28)T cells, the overexpression or the knockdown of miR 24 in resting CD8+ T cells would be necessary. However, for technical reasons this experiment is not feasible because of low histone turnover in resting CD8+ T cells.
The possibility that an attenuated DDR is independent of miR 24 can therefore not be excluded. Nevertheless, it is an intriguing hypothesis that increased miR 24 levels resulting in decreased H2AX in CD8+CD28)T cells lead to an impairment in the spreading of DNA repair foci along the chromatin (Fig. 4C). The fact that total ATM and p53 levels were unchanged in CD8+CD28)T cells but MRE11 expres sion as well as phosphorylation of ATM and p53 (Ser15) decreased in CD8+CD28)T cells upon etoposide treatment is in favor of this possi bility.
Unrepaired, clustered DNA lesions induce chromosome breakage in human cells (Asaithambyet al., 2011) and can lead to permanent cell cycle arrest or apoptosis (Mandalet al., 2011). Our data therefore argue in favor of an endogenous DNA repair deficiency in CD8+CD28)T cells leading to increased apoptosis as a consequence. To test the validity of this concept, we made use of an automated FADU assay in which basal DNA damage levels as well as the DNA repair capacity following short time exposure to etoposide were determined. Ex vivo isolated CD8+ CD28)T cells contained more damaged DNA than their CD8+CD28+ counterparts following isolation, but before stimulation. This pre existing DNA damage was increased upon etoposide treatment in a similar way as in control cells of a lower differentiation stage. After 75 min recovery time, the level of initial fluorescence signal (before etoposide treatment) was not reached in CD8+CD28)T cells suggesting a decrease in the repair capacity of these cells and consequently an accumulation of DNA dam age. These results indicate that CD8+CD28)T cells have accumulated more DNA damage upon insults in vivo, possibly as the result of a decreased DDR. This may increase their susceptibility to genotoxic stress, ultimately leading to increased apoptosis in this cell type. Whether ATM independent phosphorylation of p53 (Ser46) via HIPK2 (homeodomain interacting protein kinase 2) (D’Oraziet al., 2002; Hofmannet al., 2002) plays a role in the induction of apoptosis in CD8+CD28)T cells is presently subject of further investigation. Taken together, our results suggest that Fig. 3Initiation of DNA damage response in human CD8+CD28+and CD8+CD28 T cells following etoposide treatment. Immunofluorescence stainings of DNA (TO PRO, red) and phosphorylatedcH2AX at Ser139 (green) in CD8+CD28+and CD8+CD28 T cells after 30 minute exposure to etoposide (10lg mL1) were analyzed by confocal microscopy. One representative set of images from one donor is shown per CD8+T cell subset and treatment. The bar graph indicatescH2AX spot positive cells with the absolute number of spots in the nuclei of the cells indicated; means in % of total cells ± SEM;n= 4; **P£0.01.
CD8+CD28)T cells are short livedin vivoas they are extremely prone to die upon cellular stress.
In spite of their intrinsic susceptibility to the induction of apoptosis, CD8+CD28) T cells still accumulate during aging or in persons with persistent viral infections (Sauleet al., 2006; Brunneret al., 2011). This can be explained in two different ways: A continuous regeneration of this specific T cell subset from the CD8+CD28+T cell pool takes place as the result of persistent antigenic stimulation. This would be the case in persons with latent CMV infection or in HIV infected patients (Appay et al., 2011). Alternatively, CD8+CD28) T cells could survive for a prolonged period of time and be thus long lived in specific survival supporting niches (Beckeret al., 2005). We have recently demonstrated
that CD8+CD28)T cells are enriched in the BM and are in close contact with IL 15 producing cells (Herndler Brandstetteret al., 2011a,b). IL 15 production is high in the human BM (Herndler Brandstetter et al., 2011a,b) and still higher in old age (Herndler Brandstetteret al., 2012).
Additionally, we have shown that IL 15 strongly induces CD69 in CD8+CD28)T cells. CD69 inhibits the expression of the sphingosine 1 phosphate receptor type 1 and lymphocyte egress from lymphoid organs (Shiowet al., 2006; Maedaet al., 2010). Despite their impaired DDR, CD8+CD28)T cells display a high responsiveness to the homeostatic cytokine IL 15 (Alveset al., 2003), which increases survival, rescues T cell proliferation via the PI3K⁄AKT⁄GSK3 signaling pathway, increases anti oxidant capacity, and improves the quality of antigen driven responses (A)
(B) (C)
Fig. 4(A) DNA damage response (DDR) signaling in human CD8+CD28+and CD8+CD28 T cells after etoposide exposure. Isolated CD8+CD28+and CD8+CD28 T cells of elderly (‡65 years, left panel) and young (£30 years, right panel) donors were left as untreated control or treated with 10lg mL1etoposide for 60 min. After etoposide removal, the T cell subsets were given time to repair the induced DNA damage for 15, 30, and 60 min at 37C and 5% CO2. Regulation of essential components of the DDR machinery and apoptosis induction were analyzed by Western Blot. Data are representative of experiments with cells from elderly (n= 5) and young (n= 3) donors. (B) The graph shows the densitometrical evaluation of grayscale values. Data represent fluorescence in individual lanes as percentages of total fluorescence in the whole blot;
means ± SEM;n= 8, as there was no difference between cells from young and old donors, results from both age groups were included in the calculation; *P£0.05;
**P£0.01; CD8+CD28+(black) versus CD8+CD28 (gray). (C) Schematic representation of early DDR signaling events. The MRN nuclease⁄helicase complex, consisting of MRE11, RAD50, and NBS1, initially senses DNA strand breaks and recruits and activates monomeric ataxia telangiectasia mutated (ATM) by phosphorylation (Ser1981). At the site of DNA damage, ATM initiates the phosphorylation of H2AX (Ser139) to generatecH2AX, which elicits a sequence of signaling events that feed into a positive feedback loop to recruit more enzymes (Sengupta & Harris, 2005). Hereby,cH2AX serves as a landing pad for the retention of further essential components of the DNA repair machinery, especially MDC1 (mediator of DNA damage checkpoint 1) as well as further MRN complexes and phospho ATM molecules, to spread and reach more distal chromatin regions (West & van Attikum, 2006).
(Alonso Arias et al., 2011; Kauret al., 2011). Our results demonstrate that IL 15 signaling enables CD8+CD28)T cells to up regulate the anti apoptotic molecule BCLXLand rescue them from DNA damage induced apoptosis.
It would be interesting to investigate whether the CD8+CD28)T cells accumulating in the BM also accumulate DNA damage and a defective DNA repair mechanism. However, owing to small sample sizes, separa tion of CD8+T cell subsets from BM samples is currently not possible.
In summary, our data suggest that CD8+CD28)T cells can survive under certain circumstances in spite of their intrinsic inability to repair DNA damage and their increased capacity to undergo apoptosis. Accu mulation of this specific cell type thus greatly depends on the composition of the different homing niches of an organism. This may be of particular relevance in old age or during chronic infections, both conditions for which an increased production of inflammatory as well as homeostatic cytokines is a typical feature (Franceschiet al., 2000).
Experimental procedures
Sample preparation, purification of CD8+T-cell subsets and cell culture
Peripheral blood samples were obtained from apparently healthy elderly (‡65 years) and young (£30 years) persons who did not receive immu nomodulatory drugs nor suffered from diseases known to influence the immune system, including autoimmune diseases and⁄or cancer. All donors had given their written informed consent to participate in this study that was approved by the Ethics Committee of Innsbruck Medical University. Peripheral blood was collected in sodium heparin tubes and PBMCs were freshly isolated by Ficoll Hypaque (Amersham Biosciences, Amersham, UK) density gradient centrifugation. CD8+CD28+ and CD8+CD28)T cells were enriched from PBMCs using magnetic beads (Miltenyi Biotec, Bergisch Gladbach, Germany) in a three step MACS pro
cedure. Briefly, CD8+T cells were positively selected using anti CD8 Mul tiSort microbeads and a LS column (Miltenyi Biotec). Following removal of the CD8 MultiSort microbeads, CD8+T cells were incubated with anti CD28 allophycocyanin conjugated antibody (BD Pharmingen, Franklin Lakes, NJ, USA) and anti allophycocyanin conjugated microbeads. The CD8+CD28)T cell fraction was depleted of contaminating NK T cells Fig. 5DNA damage and repair in untreated and etoposide exposed CD8+T cell
subsets. CD8+CD28+and CD8+CD28 T cells were either left as untreated control or exposed to 10lg mL1etoposide for 60 min. After etoposide removal, both T cell subsets were given time to repair DNA damage for 15, 30, 45, 60, and 75 min at 37C and 5% CO2. The figure shows mean double stranded DNA ± SEM of six donors, independently performed and measured by an automated ‘fluorimetric detection of alkaline DNA unwinding FADU’ assay (Moreno Villanuevaet al., 2009), relative to basal CD8+CD28+T cell control. *P£0.05, **P£0.01 values versus respective controls (repair), and control values against each other (basal DNA damage, not indicated),P£0.01. While the induced DNA damage was entirely repaired after 30 min in CD8+CD28+T cells (compared to CD8+CD28+T cell control,P= n.s.), CD8+CD28 T cells remained unrepaired until the end of the observation period (compared to CD8+CD28 T cell control,P£0.05).
(A)
(B)
(C)
Fig. 6The impact of IL 15 on apoptosis and the DNA damage response in human CD8+CD28+and CD8+CD28 T cells. (A) Apoptotic CD8+CD28+and CD8+CD28 T cells after etoposide treatment. PBMCs, containing both the CD8+CD28+and the CD8+CD28 T cell subset, were either treated with etoposide only (10lg mL1) for 48 h or pre stimulated (3 h, 50 ng mL1) with IL 15 before exposing them to etoposide. Bars depict specific apoptosis as measured by 7 AAD⁄Annexin V FACS analysis; means ± SEM;n= 11; ***P£0.001; CD8+CD28+versus CD8+CD28 . (B) Apoptotic signaling under etoposide and IL 15 treatment. Isolated CD8+CD28+ and CD8+CD28 T cells were cultured for 6 h in high affinity 96 well plates coated with IL 15 (1lg mL1), while control experiments were performed in FCS blocked plates in the absence of IL 15. Etoposide (10lg mL1) was then added and the cells left to incubate for another 48 h. One representative of four Western Blots, performed with cells from four individual donors, is shown. (C) The graph shows the densitometrical evaluation of grayscale values. Data represent fluorescence in individual lanes as percentages of total fluorescence in the whole blot;
means ± SEM;n= 4; **P£0.01; CD8+CD28+(black) versus CD8+CD28 (gray).
IL 15 treatment increased BCLXLprotein levels in either CD8+T cell subset compared to control; **P£0.01.
using anti CD16 microbeads. The purity of the obtained CD8+CD28+and CD8+CD28)T cell populations was assessed by a FACSCanto II (BD Bio sciences, Franklin Lakes, NJ, USA) and yielded population homogeneities of‡90%. PBMCs and purified CD8+T cell subsets were cultured at a cell density of 2·106cells mL)1at 37C in complete RPMI 1640 medium [supplemented with 10% FCS (Sigma Aldrich, St. Louis, MO, USA) and 1% penicillin⁄streptomycin (Gibco)]. Experiments were performed with or without etoposide (10lg mL)1) or IL 15. For the stimulation of PBMCs, soluble IL 15 was used at a concentration of 50 ng mL)1, whereas isolated CD8+ T cell subsets were stimulated by adding them onto IL 15 coated high affinity 96 well plates (1lg mL)1; Greiner Bio One, Frickenhausen, Germany).
miRNA expression profiling
Two channel microarray analysis was performed as previously described (Hacklet al., 2010). In brief, locked nucleic acid probe sets (miRBase ver sion 9.2; Exiqon, Vedbaek, Denmark), comprising 559 human miRNA as well as 77 miRPlus (intellectual property of Exiqon) sequences, were spot ted on epoxy coated Nexterion glass slides (Schott AG, Mainz, Germany).
High quality total RNA was extracted from 5·106 CD8+CD28+ and CD8+CD28)T cells using TRIzol reagent (Sigma Aldrich) and Glycogen (Roche, Penzberg, Germany) as a carrier for RNA precipitation. miRNA microarray expression profiling was performed on sets of eight donor paired CD8+CD28+and CD8+CD28)T cell samples (four directlyex vivo and four cultured for 16 h to exclude a potential influence of cell culture on miRNA expression). First, 0.5lg total RNA was Cy3 and Cy5 labeled with the miRCURY LNA miRNA Array labeling kit (Exiqon). RNA samples were then hybridized for 16 h to in house spotted LNA miRNA chips using a TECAN HS 400 hybridization station (Tecan, Mannedorf, Switzer land) after which the arrays were immediately scanned by a GenePix 4000B laser scanner (Axon Instruments, Foster City, CA, USA). Cy3 dye was scanned at 532 nm and Cy5 emission was recorded at 635 nm with the resolution settings adjusted to 10lMand averaging per 1 line. Inten sity values for each spot were extracted using GenePix 4.1 software and analyzed using the Bioconductor package ‘Linear models for microarray data analysis’ under R 2.9.1. In brief, the MA transformed spot intensities of each array were background corrected (normexp algorithm) and low ess normalized (local weighted linear regression). For differential expres sion analysis, the signals of eight replicate spots for each miRNA per array were correlated and used for moderated hypothesis tests (moderatedt statistic) for the contrasts of interest between pairs of CD8+CD28+and CD8+CD28)T cells, respectively. The resultingPvalues were adjusted for multiple testing (Padj.) according to a method of Hochberg & Benjamini (1990) to control the false discovery rate (FDR). All miRNAs were then ranked in terms of their adjustedPvalues and a cut off ofP£0.05 was imposed. The respective raw data as well as processed intensity data were submitted to Array Express (http://www.ebi.ac.uk/arrayexpress) accord ing to MIAME guidelines and can be accessed under the identifiers E MEXP 2398 and E MEXP 3307.
Immunofluorescence imaging
Following MACS separation, isolated CD8+T cell subsets were washed and equilibrated in complete RPMI 1640 medium at a cell density of 2·106mL)1for 1 h at 37C. Two hundred microlitres of cell suspen sion aliquots, containing 4·105cells, were treated with 10lg mL)1 etoposide to induce DSBs for 30 min at 37C while untreated cells were used as control. Following the indicated incubation period, CD8+ T cell subsets were immediately fixed by adding 200lL CytoFix⁄Cyto
perm solution (BD Pharmingen) for 15 min at 37C. After a subsequent washing step in 1·PermWash buffer (BD Pharmingen), CD8+ T cell subsets were permeabilized for 30 min on ice in PhosFlow Perm Buffer III (BD Pharmingen). Following another washing step in 1·PermWash buffer, CD8+T cell subsets were incubated in 100lL of a primary anti body solution containing 1:400 mouse anticH2AX phospho Ser139 (Abcam, Cambridge, UK) in PermWash buffer for 1 h at 37C. Thereaf ter, washing in 1·PermWash buffer was followed by a 45 min incuba tion period at RT in 100lL of a secondary solution containing 1:200 rabbit anti mouse AlexaFlour488 conjugated Ab (Abcam) and 1:50 TO PRO solution (Invitrogen, Carlsbad, CA, USA) in 1·PermWash buffer.
After washing in 1·PermWash buffer twice, cells were resuspended in small droplets of glycerol, the cell suspension carefully mixed, trans ferred onto microscope slides, and covered with coverslips. Cells were given time to settle into the same layer by storing the slides in a vertical position for 3 h at 4C. Immunofluorescence images were taken using a lRadiance scan head (Bio Rad Laboratories, Hercules, CA, USA) attached to a Zeiss Axiophot confocal scanning microscope system and Zeiss Plan Neofluar objective lenses at 40· ⁄0.75 operated at RT and subsequently analyzed by Zeiss LaserSharp 2000 software (Carl Zeiss, Oberkochen, Germany). Experiments were performed in duplicates for each CD8+ T cell subset and independently validated for four donors.
To quantifycH2AX spot formation, the samples were screened micro scopically by two independent persons and the number of cells with clusters in their nuclei as well of the number of spots per nucleus was determined. 200 cells per sample were analyzed.
Transfection of Jurkat E6.1 cells with miR-24
As for technical reasons neither overexpression nor knockdown of miR NAs is possible in resting CD8+CD28+and CD8+CD28)T cells, we over expressed miR 24 in the human leukemic T cell lymphoblast line Jurkat E6.1. Jurkat cells were cultured at a cell density of 0.5 to 3· 106cells mL)1 in a humidified incubator operated at 37C and 5%
CO2 in complete RPMI 1640 medium (supplemented with 10% FCS and 1% penicillin⁄streptomycin) and passaged as required to maintain logarithmic growth. One hour prior to transfection, 2·105Jurkat cells in 100lL medium containing 20% FCS were transferred into 24 well plates and given time to equilibrate for the indicated time period.
750 ng (3lL of 20lM stock) miR 24 miScript miRNA mimic (Qiagen, Hilden, Germany) was diluted in 100lL medium without serum or anti biotics before adding 4lL Attractene transfection reagent (Qiagen).
Following an incubation period of 15 min to allow the formation of transfection complexes at room temperature, this suspension was added to the cells in a ratio of 1:1 and cultured for 6 h at 37C.
Another 400lL medium containing 10% FCS and antibiotics were added then, finally yielding a miRNA concentration of 100 nM. Twenty four hour post transfection 400lL fresh medium was added and the cells further cultured for another 3 days before analysis. Control Jurkat cells were treated equally but were ‘transfected’ with medium alone, were mock transfected using empty complexes (Mock), or were trans fected with nonsense miRNA (Scrambled; Qiagen).
Western Blotting
Whole cell protein lysates (TNE buffer: 50 mMTris HCl pH 8.0, 150 mM
NaCl, 0.5 mMEDTA, and 1% TritonX100, supplemented with protease and phosphatase inhibitors) of 2·106 CD8+ T cells⁄Jurkat cells per described time point and treatment were subject to Western Blot analysis (denaturing gel electrophoresis on 4 20% gradient Tris Glycine precast
polyacrylamide gels; Thermo Scientific, Waltham, MA, USA) and tested for the expression of total H2AX, phospho Ser1981 and total ATM, phos pho Ser15, phospho Ser46 and total p53, MRE11, BCLXL, PARP1⁄cleaved PARP1, and GAPDH using primary antibodies (Abcam and Cell Signaling, Danvers, MA, USA). Protein bands were visualized using horseradish per oxidase conjugated secondary Abs, ECL Western Blot substrate, or Super Signal Western Femto substrate (all Thermo Scientific), depending on the obtained signal strength, and an Alpha Innotech chemiluminescence detection unit with AlphaEaseFC software (Biozym, Hessisch Oldendorf, Germany). For H2AX expression in CD8+T cell subsets, densitometrical evaluation of obtained grayscale values normalized to GAPDH was per formed and plotted as fold regulation relative to CD8+CD28)T cells.
Automated fluorimetric detection of alkaline DNA unwinding
We assessed the efficiency of DNA repair in CD8+CD28+ and CD8+CD28)T cells by a recently developed, new generation automated version of the fluorimetric detection of alkaline DNA unwinding (FADU) assay (Moreno Villanuevaet al., 2009). This technique is based on pro gressive DNA unwinding (denaturation) under highly controlled condi tions of alkaline pH, time, and temperature. The starting points for the unwinding process are not only DSBs induced by reactive oxygen spe cies (ROS), irradiation, or chemical compounds, but also replication forks or chromosome ends. The dye SYBR Green (Invitrogen) interca lates into double stranded DNA, essentially leaving stretches of unwound DNA unaffected. The loss of the resulting fluorescence signal, compared to untreated controls, is proportional to the amount of DSBs.
The FADU assay is performed under extreme alkaline conditions that interfere with protein protein interactions and thus dissolve T and D loops present at DNA caps, which allows comparing samples with dif ferent telomere lengths. Isolated CD8+ T cell subsets were either left untreated or exposed to 10lg mL)1etoposide for 60 min as well as given time to repair the induced DNA damage after removal of etopo side for 15, 30, 45, 60, or 75 min, respectively, before performing the automated FADU assay as described elsewhere (Moreno Villanueva et al., 2009). Samples were analyzed in a 96 well plate fluorescence reader at 492 nm excitation⁄520 nm emission immediately after SYBR Green addition. Experiments were performed with three technical repli cates per sample and the average calculated. Values from six paired samples of CD8+CD28+and CD8+CD28)T cells obtained from six indi vidual donors, performed on three independent runs, were plotted onto a scale representing means of double stranded DNA compared to basal CD8+CD28+ T cell levels. The recovery times after DNA damage were calculated using two tailed Student’sttest.
Quantitative RT–PCR
Total RNA was extracted from untreated and etoposide treated CD8+CD28+ and CD8+CD28)T cells⁄Jurkat cells and cDNA was syn thesized by applying the miRCURY LNA Universal real time (RT) system (Exiqon) for evaluation of miR 24 expression. Quantitative RT PCR experi ments were performed using the LightCycler II 480 System (Roche Diag nostics, Risch, Switzerland) and GAPDH (fwd: GAGTCAACGGATTTGG TCGT, rvs: GATCTCGCTCCTGGAAGATG) or U6 miRNA as housekeeping genes for normalization purposes and the relative quantification of target genes. The amplification protocol used was as follows: initial incubation at 95C for 8 min followed by 40 amplification cycles at 95C for 10 s and 60C for 60 s according to the manufacturer’s instructions. Quality control of amplified gene targets was performed by assessing the melting
peak profiles of obtained PCR products. GAPDH normalized threshold cycle (CT) values (DCT CTgene CTGAPDH) were used to calculate the relative gene expression compared to control samples (2)DDCT).
Assessment of apoptosis by Annexin V⁄7-AAD staining Viable cells were determined counting Annexin V⁄7 AAD double nega tive cells by FACS analysis (BD FACS Canto II with BD FACS Diva software, 40 000 events analyzed). Experiments were performed on PBMCs, count erstained with anti CD28 APC, anti CD3 APC Cy7, and anti CD8 PE (BD Pharmingen) as well as on isolated CD8+T cell subsets. Percent specific apoptosis over background was calculated by normalizing against respec tive controls.
Acknowledgments
This work was supported by the Austrian Science Fund (S9308-B05 to BGL; Y212-B13 to AV; S93-06 to JG) and by the EU-funded Net- work of Excellence LifeSpan (FP6 036894). JG is also supported by GEN-AU Project 820982 ‘Non-coding RNAs’, by CE.R.I.E.S., and the Herzfelder’sche Familienstiftung. DHB was supported by a ‘Future Leaders of Ageing Research in Europe(FLARE)’ fellowship funded by the Austrian Federal Ministry of Science and Research; MH is sup- ported by BOKU-DOC.
Author contributions
SB, DHB, and CA designed the experiments, planned the study, per- formed experiments, and interpreted the data; MH and JG spotted LNA-based miRNA arrays and performed statistical analysis on array data; GJW and AV provided invaluable expertize on T-cell apoptosis;
MMW and AB invented, designed, and constructed the automated FADU assay and allowed SB to perform experiments on this device;
SB and BGL wrote the manuscript.
References
Alonso Arias R, Moro Garcia MA, Vidal Castineira JR, Solano Jaurrieta JJ, Suarez Garcia FM, Coto E, Lopez Larrea C (2011) IL 15 preferentially enhances func tional properties and antigen specific responses of CD4+CD28(null) compared to CD4+CD28+ T cells.Aging Cell10, 844 852.
Alves NL, Hooibrink B, Arosa FA, van Lier RA (2003) IL 15 induces antigen inde pendent expansion and differentiation of human naive CD8+ T cellsin vitro.
Blood102, 2541 2546.
Appay V, Fastenackels S, Katlama C, Ait Mohand H, Schneider L, Guihot A, Kel ler M, Grubeck Loebenstein B, Simon A, Lambotte O, Hunt PW, Deeks SG, Costagliola D, Autran B, Sauce D (2011) Old age and anti CMV immunity are associated with altered T cell reconstitution in HIV 1 infected patients.AIDS 25, 1813 1822.
Arnold CR, Wolf J, Brunner S, Herndler Brandstetter D, Grubeck Loebenstein B (2011) Gain and loss of T cell subsets in old age age related reshaping of the T cell repertoire.J. Clin. Immunol.31, 137 146.
Asaithamby A, Hu B, Chen DJ (2011) Unrepaired clustered DNA lesions induce chro mosome breakage in human cells.Proc. Natl Acad. Sci. USA108, 8293 8298.
Baldwin EL, Osheroff N (2005) Etoposide, topoisomerase II and cancer.Curr.
Med. Chem. Anticancer Agents5, 363 372.
Baltimore D, Boldin MP, O’Connell RM, Rao DS, Taganov KD (2008) MicroRNAs:
new regulators of immune cell development and function.Nat. Immunol.9, 839 845.
Becker TC, Coley SM, Wherry EJ, Ahmed R (2005) Bone marrow is a preferred site for homeostatic proliferation of memory CD8 T cells. J. Immunol.174, 1269 1273.
Brunner S, Herndler Brandstetter D, Weinberger B, Grubeck Loebenstein B (2011) Persistent viral infections and immune aging. Ageing Res. Rev. 10, 362 369.
D’Orazi G, Cecchinelli B, Bruno T, Manni I, Higashimoto Y, Saito S, Gostissa M, Coen S, Marchetti A, Del Sal G, Piaggio G, Fanciulli M, Appella E, Soddu S (2002) Homeodomain interacting protein kinase 2 phosphorylates p53 at Ser 46 and mediates apoptosis.Nat. Cell Biol.4, 11 19.
Effros RB, Dagarag M, Spaulding C, Man J (2005) The role of CD8+ T cell repli cative senescence in human aging.Immunol. Rev.205, 147 157.
Franceschi C, Bonafe M, Valensin S, Olivieri F, De Luca M, Ottaviani E, De Bene dictis G (2000) Inflamm aging. An evolutionary perspective on immunosenes cence.Ann. N. Y. Acad. Sci.908, 244 254.
Goronzy JJ, Fulbright JW, Crowson CS, Poland GA, O’Fallon WM, Weyand CM (2001) Value of immunological markers in predicting responsiveness to influ enza vaccination in elderly individuals.J. Virol.75, 12182 12187.
Grillari J, Grillari Voglauer R (2010) Novel modulators of senescence, aging, and longevity: small non coding RNAs enter the stage.Exp. Gerontol.45, 302 311.
Grillari J, Hackl M, Grillari Voglauer R (2010) miR 17 92 cluster: ups and downs in cancer and aging.Biogerontology11, 501 506.
van Haaften G, Agami R (2010) Tumorigenicity of the miR 17 92 cluster distilled.
Genes Dev.24, 1 4.
Herndler Brandstetter D, Landgraf K, Tzankov A, Jenewein B, Brunauer R, Laschober GT, Parson W, Kloss F, Gassner R, Lepperdinger G, Grubeck Loebenstein B (2012) The impact of aging on memory T cell phenotype and function in the human bone marrow.J. Leuk. Biol.91, 197 205.
Hackl M, Brunner S, Fortschegger K, Schreiner C, Micutkova L, Muck C, Lascho ber GT, Lepperdinger G, Sampson N, Berger P, Herndler Brandstetter D, Wie ser M, Kuhnel H, Strasser A, Rinnerthaler M, Breitenbach M, Mildner M, Eckhart L, Tschachler E, Trost A, Bauer JW, Papak C, Trajanoski Z, Scheideler M, Grillari Voglauer R, Grubeck Loebenstein B, Jansen Durr P, Grillari J (2010) miR 17, miR 19b, miR 20a, and miR 106a are down regulated in human aging.Aging Cell9, 291 296.
Herndler Brandstetter D, Brunner S, Weiskopf D, van Rijn R, Landgraf K, Dejaco C, Duftner C, Schirmer M, Kloss F, Gassner R, Lepperdinger G, Grubeck Loe benstein B (2011a) Post thymic regulation of CD5 levels in human memory T cells is inversely associated with the strength of responsiveness to interleukin 15.Hum. Immunol.72, 627 631.
Herndler Brandstetter D, Landgraf K, Jenewein B, Tzankov A, Brunauer R, Brunner S, Parson W, Kloss F, Gassner R, Lepperdinger G, Grubeck Loebenstein B (2011b) Human bone marrow hosts polyfunctional memory CD4+ and CD8+ T cells with close contact to IL 15 producing cells.J. Immunol.186, 6965 6971.
Hochberg Y, Benjamini Y (1990) More powerful procedures for multiple signifi cance testing.Stat. Med.9, 811 818.
Hofmann TG, Moller A, Sirma H, Zentgraf H, Taya Y, Droge W, Will H, Schmitz ML (2002) Regulation of p53 activity by its interaction with homeodomain interacting protein kinase 2.Nat. Cell Biol.4, 1 10.
Ivanovska I, Ball AS, Diaz RL, Magnus JF, Kibukawa M, Schelter JM, Kobayashi SV, Lim L, Burchard J, Jackson AL, Linsley PS, Cleary MA (2008) MicroRNAs in the miR 106b family regulate p21⁄CDKN1A and promote cell cycle progres sion.Mol. Cell. Biol.28, 2167 2174.
Kaur N, Naga OS, Norell H, Al Khami AA, Scheffel MJ, Chakraborty NG, Voelkel Johnson C, Mukherji B, Mehrotra S (2011) T cells expanded in presence of IL 15 exhibit increased antioxidant capacity and innate effector molecules.
Cytokine55, 307 317.
Lal A, Pan Y, Navarro F, Dykxhoorn DM, Moreau L, Meire E, Bentwich Z, Lieber man J, Chowdhury D (2009) miR 24 mediated downregulation of H2AX
suppresses DNA repair in terminally differentiated blood cells.Nat. Struct. Mol.
Biol.16, 492 498.
Maeda Y, Seki N, Sato N, Sugahara K, Chiba K (2010) Sphingosine 1 phosphate receptor type 1 regulates egress of mature T cells from mouse bone marrow.
Int. Immunol.22, 515 525.
Mandal PK, Blanpain C, Rossi DJ (2011) DNA damage response in adult stem cells: pathways and consequences.Nat. Rev. Mol. Cell Biol.12, 198 202.
Moreno Villanueva M, Pfeiffer R, Sindlinger T, Leake A, Muller M, Kirkwood TB, Burkle A (2009) A modified and automated version of the ‘Fluorimetric Detec tion of Alkaline DNA Unwinding’ method to quantify formation and repair of DNA strand breaks.BMC Biotechnol.9, 39.
Pawelec G, Koch S, Gouttefangeas C, Wikby A (2006) Immunorejuvenation in the elderly.Rejuvenation Res.9, 111 116.
Saule P, Trauet J, Dutriez V, Lekeux V, Dessaint JP, Labalette M (2006) Accumulation of memory T cells from childhood to old age: central and effector memory cells in CD4(+) versus effector memory and terminally differentiated memory cells in CD8(+) compartment.Mech. Ageing Dev.127, 274 281.
Saurwein Teissl M, Lung TL, Marx F, Gschosser C, Asch E, Blasko I, Parson W, Bock G, Schonitzer D, Trannoy E, Grubeck Loebenstein B (2002) Lack of anti body production following immunization in old age: association with CD8(+)CD28( ) T cell clonal expansions and an imbalance in the production of Th1 and Th2 cytokines.J. Immunol.168, 5893 5899.
Scarpaci S, Frasca D, Barattini P, Guidi L, Doria G (2003) DNA damage recognition and repair capacities in human naive and memory T cells from peripheral blood of young and elderly subjects.Mech. Ageing Dev.124, 517 524.
Seluanov A, Mittelman D, Pereira Smith OM, Wilson JH, Gorbunova V (2004) DNA end joining becomes less efficient and more error prone during cellular senescence.Proc. Natl Acad. Sci. USA101, 7624 7629.
Sengupta S, Harris CC (2005) p53: traffic cop at the crossroads of DNA repair and recombination.Nat. Rev. Mol. Cell Biol.6, 44 55.
Shiow LR, Rosen DB, Brdickova N, Xu Y, An J, Lanier LL, Cyster JG, Matloubian M (2006) CD69 acts downstream of interferon alpha⁄beta to inhibit S1P1 and lymphocyte egress from lymphoid organs.Nature440, 540 544.
Spaulding C, Guo W, Effros RB (1999) Resistance to apoptosis in human CD8+ T cells that reach replicative senescence after multiple rounds of antigen specific proliferation.Exp. Gerontol.34, 633 644.
Srivastava N, Manvati S, Srivastava A, Pal R, Kalaiarasan P, Chattopadhyay S, Gochhait S, Dua R, Bamezai RN (2011) miR 24 2 controls H2AFX expression regardless of gene copy number alteration and induces apoptosis by targeting antiapoptotic gene BCL 2: a potential for therapeutic intervention.Breast Can cer Res.13, R39.
Steinmann GG, Klaus B, Muller Hermelink HK (1985) The involution of the age ing human thymic epithelium is independent of puberty. A morphometric study.Scand. J. Immunol.22, 563 575.
Wallace DL, Masters JE, De Lara CM, Henson SM, Worth A, Zhang Y, Kumar SR, Beverley PC, Akbar AN, Macallan DC (2011) Human cytomegalovirus specific CD8(+) T cell expansions contain long lived cells that retain functional capacity in both young and elderly subjects.Immunology132, 27 38.
West AG, van Attikum H (2006) Chromatin at the crossroads. Meeting on signal ling to chromatin epigenetics.EMBO Rep.7, 1206 1210.
Wikby A, Ferguson F, Forsey R, Thompson J, Strindhall J, Lofgren S, Nilsson BO, Ernerudh J, Pawelec G, Johansson B (2005) An immune risk phenotype, cogni tive impairment, and survival in very late life: impact of allostatic load in Swed ish octogenarian and nonagenarian humans.J. Gerontol. A Biol. Sci. Med. Sci.
60, 556 565.
